Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "Fixes for kvm on x86:

   - new selftests

   - fixes for migration with HyperV re-enlightenment enabled

   - fix RCU/SRCU usage

   - fixes for local_irq_restore misuse false positive"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  documentation/kvm: additional explanations on KVM_SET_BOOT_CPU_ID
  x86/kvm: Fix broken irq restoration in kvm_wait
  KVM: X86: Fix missing local pCPU when executing wbinvd on all dirty pCPUs
  KVM: x86: Protect userspace MSR filter with SRCU, and set atomically-ish
  selftests: kvm: add set_boot_cpu_id test
  selftests: kvm: add _vm_ioctl
  selftests: kvm: add get_msr_index_features
  selftests: kvm: Add basic Hyper-V clocksources tests
  KVM: x86: hyper-v: Don't touch TSC page values when guest opted for re-enlightenment
  KVM: x86: hyper-v: Track Hyper-V TSC page status
  KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs
  KVM: x86: hyper-v: Limit guest to writing zero to HV_X64_MSR_TSC_EMULATION_STATUS
  KVM: x86/mmu: Store the address space ID in the TDP iterator
  KVM: x86/mmu: Factor out tdp_iter_return_to_root
  KVM: x86/mmu: Fix RCU usage when atomically zapping SPTEs
  KVM: x86/mmu: Fix RCU usage in handle_removed_tdp_mmu_page

Documentation/virt/kvm/api.rst

@@ -1495,7 +1495,8 @@ Fails if any VCPU has already been created.
 
 Define which vcpu is the Bootstrap Processor (BSP).  Values are the same
 as the vcpu id in KVM_CREATE_VCPU.  If this ioctl is not called, the default
-is vcpu 0.
+is vcpu 0. This ioctl has to be called before vcpu creation,
+otherwise it will return EBUSY error.
 
 
 4.42 KVM_GET_XSAVE
@@ -4806,8 +4807,10 @@ If an MSR access is not permitted through the filtering, it generates a
 allows user space to deflect and potentially handle various MSR accesses
 into user space.
 
-If a vCPU is in running state while this ioctl is invoked, the vCPU may
-experience inconsistent filtering behavior on MSR accesses.
+Note, invoking this ioctl with a vCPU is running is inherently racy.  However,
+KVM does guarantee that vCPUs will see either the previous filter or the new
+filter, e.g. MSRs with identical settings in both the old and new filter will
+have deterministic behavior.
 
 4.127 KVM_XEN_HVM_SET_ATTR
 --------------------------

arch/x86/include/asm/kvm_host.h

@@ -884,12 +884,29 @@ struct kvm_hv_syndbg {
 	u64 options;
 };
 
+/* Current state of Hyper-V TSC page clocksource */
+enum hv_tsc_page_status {
+	/* TSC page was not set up or disabled */
+	HV_TSC_PAGE_UNSET = 0,
+	/* TSC page MSR was written by the guest, update pending */
+	HV_TSC_PAGE_GUEST_CHANGED,
+	/* TSC page MSR was written by KVM userspace, update pending */
+	HV_TSC_PAGE_HOST_CHANGED,
+	/* TSC page was properly set up and is currently active  */
+	HV_TSC_PAGE_SET,
+	/* TSC page is currently being updated and therefore is inactive */
+	HV_TSC_PAGE_UPDATING,
+	/* TSC page was set up with an inaccessible GPA */
+	HV_TSC_PAGE_BROKEN,
+};
+
 /* Hyper-V emulation context */
 struct kvm_hv {
 	struct mutex hv_lock;
 	u64 hv_guest_os_id;
 	u64 hv_hypercall;
 	u64 hv_tsc_page;
+	enum hv_tsc_page_status hv_tsc_page_status;
 
 	/* Hyper-v based guest crash (NT kernel bugcheck) parameters */
 	u64 hv_crash_param[HV_X64_MSR_CRASH_PARAMS];
@@ -931,6 +948,12 @@ enum kvm_irqchip_mode {
 	KVM_IRQCHIP_SPLIT,        /* created with KVM_CAP_SPLIT_IRQCHIP */
 };
 
+struct kvm_x86_msr_filter {
+	u8 count;
+	bool default_allow:1;
+	struct msr_bitmap_range ranges[16];
+};
+
 #define APICV_INHIBIT_REASON_DISABLE    0
 #define APICV_INHIBIT_REASON_HYPERV     1
 #define APICV_INHIBIT_REASON_NESTED     2
@@ -1025,16 +1048,11 @@ struct kvm_arch {
 	bool guest_can_read_msr_platform_info;
 	bool exception_payload_enabled;
 
+	bool bus_lock_detection_enabled;
+
 	/* Deflect RDMSR and WRMSR to user space when they trigger a #GP */
 	u32 user_space_msr_mask;
-
-	struct {
-		u8 count;
-		bool default_allow:1;
-		struct msr_bitmap_range ranges[16];
-	} msr_filter;
-
-	bool bus_lock_detection_enabled;
+	struct kvm_x86_msr_filter __rcu *msr_filter;
 
 	struct kvm_pmu_event_filter __rcu *pmu_event_filter;
 	struct task_struct *nx_lpage_recovery_thread;

arch/x86/kernel/kvm.c

@@ -836,28 +836,25 @@ static void kvm_kick_cpu(int cpu)
 
 static void kvm_wait(u8 *ptr, u8 val)
 {
-	unsigned long flags;
-
 	if (in_nmi())
 		return;
 
-	local_irq_save(flags);
-
-	if (READ_ONCE(*ptr) != val)
-		goto out;
-
 	/*
 	 * halt until it's our turn and kicked. Note that we do safe halt
 	 * for irq enabled case to avoid hang when lock info is overwritten
 	 * in irq spinlock slowpath and no spurious interrupt occur to save us.
 	 */
-	if (arch_irqs_disabled_flags(flags))
-		halt();
-	else
-		safe_halt();
+	if (irqs_disabled()) {
+		if (READ_ONCE(*ptr) == val)
+			halt();
+	} else {
+		local_irq_disable();
 
-out:
-	local_irq_restore(flags);
+		if (READ_ONCE(*ptr) == val)
+			safe_halt();
+
+		local_irq_enable();
+	}
 }
 
 #ifdef CONFIG_X86_32

arch/x86/kvm/hyperv.c

@@ -520,10 +520,10 @@ static u64 get_time_ref_counter(struct kvm *kvm)
 	u64 tsc;
 
 	/*
-	 * The guest has not set up the TSC page or the clock isn't
-	 * stable, fall back to get_kvmclock_ns.
+	 * Fall back to get_kvmclock_ns() when TSC page hasn't been set up,
+	 * is broken, disabled or being updated.
 	 */
-	if (!hv->tsc_ref.tsc_sequence)
+	if (hv->hv_tsc_page_status != HV_TSC_PAGE_SET)
 		return div_u64(get_kvmclock_ns(kvm), 100);
 
 	vcpu = kvm_get_vcpu(kvm, 0);
@@ -1077,6 +1077,21 @@ static bool compute_tsc_page_parameters(struct pvclock_vcpu_time_info *hv_clock,
 	return true;
 }
 
+/*
+ * Don't touch TSC page values if the guest has opted for TSC emulation after
+ * migration. KVM doesn't fully support reenlightenment notifications and TSC
+ * access emulation and Hyper-V is known to expect the values in TSC page to
+ * stay constant before TSC access emulation is disabled from guest side
+ * (HV_X64_MSR_TSC_EMULATION_STATUS). KVM userspace is expected to preserve TSC
+ * frequency and guest visible TSC value across migration (and prevent it when
+ * TSC scaling is unsupported).
+ */
+static inline bool tsc_page_update_unsafe(struct kvm_hv *hv)
+{
+	return (hv->hv_tsc_page_status != HV_TSC_PAGE_GUEST_CHANGED) &&
+		hv->hv_tsc_emulation_control;
+}
+
 void kvm_hv_setup_tsc_page(struct kvm *kvm,
 			   struct pvclock_vcpu_time_info *hv_clock)
 {
@@ -1087,7 +1102,8 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	BUILD_BUG_ON(sizeof(tsc_seq) != sizeof(hv->tsc_ref.tsc_sequence));
 	BUILD_BUG_ON(offsetof(struct ms_hyperv_tsc_page, tsc_sequence) != 0);
 
-	if (!(hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE))
+	if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
+	    hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET)
 		return;
 
 	mutex_lock(&hv->hv_lock);
@@ -1101,7 +1117,15 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	 */
 	if (unlikely(kvm_read_guest(kvm, gfn_to_gpa(gfn),
 				    &tsc_seq, sizeof(tsc_seq))))
+		goto out_err;
+
+	if (tsc_seq && tsc_page_update_unsafe(hv)) {
+		if (kvm_read_guest(kvm, gfn_to_gpa(gfn), &hv->tsc_ref, sizeof(hv->tsc_ref)))
+			goto out_err;
+
+		hv->hv_tsc_page_status = HV_TSC_PAGE_SET;
 		goto out_unlock;
+	}
 
 	/*
 	 * While we're computing and writing the parameters, force the
@@ -1110,15 +1134,15 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	hv->tsc_ref.tsc_sequence = 0;
 	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
 			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
-		goto out_unlock;
+		goto out_err;
 
 	if (!compute_tsc_page_parameters(hv_clock, &hv->tsc_ref))
-		goto out_unlock;
+		goto out_err;
 
 	/* Ensure sequence is zero before writing the rest of the struct.  */
 	smp_wmb();
 	if (kvm_write_guest(kvm, gfn_to_gpa(gfn), &hv->tsc_ref, sizeof(hv->tsc_ref)))
-		goto out_unlock;
+		goto out_err;
 
 	/*
 	 * Now switch to the TSC page mechanism by writing the sequence.
@@ -1131,8 +1155,45 @@ void kvm_hv_setup_tsc_page(struct kvm *kvm,
 	smp_wmb();
 
 	hv->tsc_ref.tsc_sequence = tsc_seq;
-	kvm_write_guest(kvm, gfn_to_gpa(gfn),
-			&hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence));
+	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
+			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
+		goto out_err;
+
+	hv->hv_tsc_page_status = HV_TSC_PAGE_SET;
+	goto out_unlock;
+
+out_err:
+	hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
+out_unlock:
+	mutex_unlock(&hv->hv_lock);
+}
+
+void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
+{
+	struct kvm_hv *hv = to_kvm_hv(kvm);
+	u64 gfn;
+
+	if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
+	    hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET ||
+	    tsc_page_update_unsafe(hv))
+		return;
+
+	mutex_lock(&hv->hv_lock);
+
+	if (!(hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE))
+		goto out_unlock;
+
+	/* Preserve HV_TSC_PAGE_GUEST_CHANGED/HV_TSC_PAGE_HOST_CHANGED states */
+	if (hv->hv_tsc_page_status == HV_TSC_PAGE_SET)
+		hv->hv_tsc_page_status = HV_TSC_PAGE_UPDATING;
+
+	gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
+
+	hv->tsc_ref.tsc_sequence = 0;
+	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
+			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
+		hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
+
 out_unlock:
 	mutex_unlock(&hv->hv_lock);
 }
@@ -1193,8 +1254,15 @@ static int kvm_hv_set_msr_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data,
 	}
 	case HV_X64_MSR_REFERENCE_TSC:
 		hv->hv_tsc_page = data;
-		if (hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE)
+		if (hv->hv_tsc_page & HV_X64_MSR_TSC_REFERENCE_ENABLE) {
+			if (!host)
+				hv->hv_tsc_page_status = HV_TSC_PAGE_GUEST_CHANGED;
+			else
+				hv->hv_tsc_page_status = HV_TSC_PAGE_HOST_CHANGED;
 			kvm_make_request(KVM_REQ_MASTERCLOCK_UPDATE, vcpu);
+		} else {
+			hv->hv_tsc_page_status = HV_TSC_PAGE_UNSET;
+		}
 		break;
 	case HV_X64_MSR_CRASH_P0 ... HV_X64_MSR_CRASH_P4:
 		return kvm_hv_msr_set_crash_data(kvm,
@@ -1229,6 +1297,9 @@ static int kvm_hv_set_msr_pw(struct kvm_vcpu *vcpu, u32 msr, u64 data,
 		hv->hv_tsc_emulation_control = data;
 		break;
 	case HV_X64_MSR_TSC_EMULATION_STATUS:
+		if (data && !host)
+			return 1;
+
 		hv->hv_tsc_emulation_status = data;
 		break;
 	case HV_X64_MSR_TIME_REF_COUNT:

arch/x86/kvm/hyperv.h

@@ -133,6 +133,7 @@ void kvm_hv_process_stimers(struct kvm_vcpu *vcpu);
 
 void kvm_hv_setup_tsc_page(struct kvm *kvm,
 			   struct pvclock_vcpu_time_info *hv_clock);
+void kvm_hv_invalidate_tsc_page(struct kvm *kvm);
 
 void kvm_hv_init_vm(struct kvm *kvm);
 void kvm_hv_destroy_vm(struct kvm *kvm);

arch/x86/kvm/mmu/mmu_internal.h

@@ -78,6 +78,11 @@ static inline struct kvm_mmu_page *sptep_to_sp(u64 *sptep)
 	return to_shadow_page(__pa(sptep));
 }
 
+static inline int kvm_mmu_page_as_id(struct kvm_mmu_page *sp)
+{
+	return sp->role.smm ? 1 : 0;
+}
+
 static inline bool kvm_vcpu_ad_need_write_protect(struct kvm_vcpu *vcpu)
 {
 	/*

arch/x86/kvm/mmu/tdp_iter.c

@@ -20,6 +20,21 @@ static gfn_t round_gfn_for_level(gfn_t gfn, int level)
 	return gfn & -KVM_PAGES_PER_HPAGE(level);
 }
 
+/*
+ * Return the TDP iterator to the root PT and allow it to continue its
+ * traversal over the paging structure from there.
+ */
+void tdp_iter_restart(struct tdp_iter *iter)
+{
+	iter->yielded_gfn = iter->next_last_level_gfn;
+	iter->level = iter->root_level;
+
+	iter->gfn = round_gfn_for_level(iter->next_last_level_gfn, iter->level);
+	tdp_iter_refresh_sptep(iter);
+
+	iter->valid = true;
+}
+
 /*
  * Sets a TDP iterator to walk a pre-order traversal of the paging structure
  * rooted at root_pt, starting with the walk to translate next_last_level_gfn.
@@ -31,16 +46,12 @@ void tdp_iter_start(struct tdp_iter *iter, u64 *root_pt, int root_level,
 	WARN_ON(root_level > PT64_ROOT_MAX_LEVEL);
 
 	iter->next_last_level_gfn = next_last_level_gfn;
-	iter->yielded_gfn = iter->next_last_level_gfn;
 	iter->root_level = root_level;
 	iter->min_level = min_level;
-	iter->level = root_level;
-	iter->pt_path[iter->level - 1] = (tdp_ptep_t)root_pt;
-
-	iter->gfn = round_gfn_for_level(iter->next_last_level_gfn, iter->level);
-	tdp_iter_refresh_sptep(iter);
+	iter->pt_path[iter->root_level - 1] = (tdp_ptep_t)root_pt;
+	iter->as_id = kvm_mmu_page_as_id(sptep_to_sp(root_pt));
 
-	iter->valid = true;
+	tdp_iter_restart(iter);
 }
 
 /*
@@ -159,8 +170,3 @@ void tdp_iter_next(struct tdp_iter *iter)
 	iter->valid = false;
 }
 
-tdp_ptep_t tdp_iter_root_pt(struct tdp_iter *iter)
-{
-	return iter->pt_path[iter->root_level - 1];
-}
-

arch/x86/kvm/mmu/tdp_iter.h

@@ -36,6 +36,8 @@ struct tdp_iter {
 	int min_level;
 	/* The iterator's current level within the paging structure */
 	int level;
+	/* The address space ID, i.e. SMM vs. regular. */
+	int as_id;
 	/* A snapshot of the value at sptep */
 	u64 old_spte;
 	/*
@@ -62,6 +64,6 @@ tdp_ptep_t spte_to_child_pt(u64 pte, int level);
 void tdp_iter_start(struct tdp_iter *iter, u64 *root_pt, int root_level,
 		    int min_level, gfn_t next_last_level_gfn);
 void tdp_iter_next(struct tdp_iter *iter);
-tdp_ptep_t tdp_iter_root_pt(struct tdp_iter *iter);
+void tdp_iter_restart(struct tdp_iter *iter);
 
 #endif /* __KVM_X86_MMU_TDP_ITER_H */

arch/x86/kvm/mmu/tdp_mmu.c

@@ -203,11 +203,6 @@ static void handle_changed_spte(struct kvm *kvm, int as_id, gfn_t gfn,
 				u64 old_spte, u64 new_spte, int level,
 				bool shared);
 
-static int kvm_mmu_page_as_id(struct kvm_mmu_page *sp)
-{
-	return sp->role.smm ? 1 : 0;
-}
-
 static void handle_changed_spte_acc_track(u64 old_spte, u64 new_spte, int level)
 {
 	bool pfn_changed = spte_to_pfn(old_spte) != spte_to_pfn(new_spte);
@@ -301,11 +296,16 @@ static void tdp_mmu_unlink_page(struct kvm *kvm, struct kvm_mmu_page *sp,
  *
  * Given a page table that has been removed from the TDP paging structure,
  * iterates through the page table to clear SPTEs and free child page tables.
+ *
+ * Note that pt is passed in as a tdp_ptep_t, but it does not need RCU
+ * protection. Since this thread removed it from the paging structure,
+ * this thread will be responsible for ensuring the page is freed. Hence the
+ * early rcu_dereferences in the function.
  */
-static void handle_removed_tdp_mmu_page(struct kvm *kvm, u64 *pt,
+static void handle_removed_tdp_mmu_page(struct kvm *kvm, tdp_ptep_t pt,
 					bool shared)
 {
-	struct kvm_mmu_page *sp = sptep_to_sp(pt);
+	struct kvm_mmu_page *sp = sptep_to_sp(rcu_dereference(pt));
 	int level = sp->role.level;
 	gfn_t base_gfn = sp->gfn;
 	u64 old_child_spte;
@@ -318,7 +318,7 @@ static void handle_removed_tdp_mmu_page(struct kvm *kvm, u64 *pt,
 	tdp_mmu_unlink_page(kvm, sp, shared);
 
 	for (i = 0; i < PT64_ENT_PER_PAGE; i++) {
-		sptep = pt + i;
+		sptep = rcu_dereference(pt) + i;
 		gfn = base_gfn + (i * KVM_PAGES_PER_HPAGE(level - 1));
 
 		if (shared) {
@@ -492,10 +492,6 @@ static inline bool tdp_mmu_set_spte_atomic(struct kvm *kvm,
 					   struct tdp_iter *iter,
 					   u64 new_spte)
 {
-	u64 *root_pt = tdp_iter_root_pt(iter);
-	struct kvm_mmu_page *root = sptep_to_sp(root_pt);
-	int as_id = kvm_mmu_page_as_id(root);
-
 	lockdep_assert_held_read(&kvm->mmu_lock);
 
 	/*
@@ -509,8 +505,8 @@ static inline bool tdp_mmu_set_spte_atomic(struct kvm *kvm,
 		      new_spte) != iter->old_spte)
 		return false;
 
-	handle_changed_spte(kvm, as_id, iter->gfn, iter->old_spte, new_spte,
-			    iter->level, true);
+	handle_changed_spte(kvm, iter->as_id, iter->gfn, iter->old_spte,
+			    new_spte, iter->level, true);
 
 	return true;
 }
@@ -538,7 +534,7 @@ static inline bool tdp_mmu_zap_spte_atomic(struct kvm *kvm,
 	 * here since the SPTE is going from non-present
 	 * to non-present.
 	 */
-	WRITE_ONCE(*iter->sptep, 0);
+	WRITE_ONCE(*rcu_dereference(iter->sptep), 0);
 
 	return true;
 }
@@ -564,10 +560,6 @@ static inline void __tdp_mmu_set_spte(struct kvm *kvm, struct tdp_iter *iter,
 				      u64 new_spte, bool record_acc_track,
 				      bool record_dirty_log)
 {
-	tdp_ptep_t root_pt = tdp_iter_root_pt(iter);
-	struct kvm_mmu_page *root = sptep_to_sp(root_pt);
-	int as_id = kvm_mmu_page_as_id(root);
-
 	lockdep_assert_held_write(&kvm->mmu_lock);
 
 	/*
@@ -581,13 +573,13 @@ static inline void __tdp_mmu_set_spte(struct kvm *kvm, struct tdp_iter *iter,
 
 	WRITE_ONCE(*rcu_dereference(iter->sptep), new_spte);
 
-	__handle_changed_spte(kvm, as_id, iter->gfn, iter->old_spte, new_spte,
-			      iter->level, false);
+	__handle_changed_spte(kvm, iter->as_id, iter->gfn, iter->old_spte,
+			      new_spte, iter->level, false);
 	if (record_acc_track)
 		handle_changed_spte_acc_track(iter->old_spte, new_spte,
 					      iter->level);
 	if (record_dirty_log)
-		handle_changed_spte_dirty_log(kvm, as_id, iter->gfn,
+		handle_changed_spte_dirty_log(kvm, iter->as_id, iter->gfn,
 					      iter->old_spte, new_spte,
 					      iter->level);
 }
@@ -659,9 +651,7 @@ static inline bool tdp_mmu_iter_cond_resched(struct kvm *kvm,
 
 		WARN_ON(iter->gfn > iter->next_last_level_gfn);
 
-		tdp_iter_start(iter, iter->pt_path[iter->root_level - 1],
-			       iter->root_level, iter->min_level,
-			       iter->next_last_level_gfn);
+		tdp_iter_restart(iter);
 
 		return true;
 	}

arch/x86/kvm/x86.c

@@ -1526,35 +1526,44 @@ EXPORT_SYMBOL_GPL(kvm_enable_efer_bits);
 
 bool kvm_msr_allowed(struct kvm_vcpu *vcpu, u32 index, u32 type)
 {
+	struct kvm_x86_msr_filter *msr_filter;
+	struct msr_bitmap_range *ranges;
 	struct kvm *kvm = vcpu->kvm;
-	struct msr_bitmap_range *ranges = kvm->arch.msr_filter.ranges;
-	u32 count = kvm->arch.msr_filter.count;
-	u32 i;
-	bool r = kvm->arch.msr_filter.default_allow;
+	bool allowed;
 	int idx;
+	u32 i;
 
-	/* MSR filtering not set up or x2APIC enabled, allow everything */
-	if (!count || (index >= 0x800 && index <= 0x8ff))
+	/* x2APIC MSRs do not support filtering. */
+	if (index >= 0x800 && index <= 0x8ff)
 		return true;
 
-	/* Prevent collision with set_msr_filter */
 	idx = srcu_read_lock(&kvm->srcu);
 
-	for (i = 0; i < count; i++) {
+	msr_filter = srcu_dereference(kvm->arch.msr_filter, &kvm->srcu);
+	if (!msr_filter) {
+		allowed = true;
+		goto out;
+	}
+
+	allowed = msr_filter->default_allow;
+	ranges = msr_filter->ranges;
+
+	for (i = 0; i < msr_filter->count; i++) {
 		u32 start = ranges[i].base;
 		u32 end = start + ranges[i].nmsrs;
 		u32 flags = ranges[i].flags;
 		unsigned long *bitmap = ranges[i].bitmap;
 
 		if ((index >= start) && (index < end) && (flags & type)) {
-			r = !!test_bit(index - start, bitmap);
+			allowed = !!test_bit(index - start, bitmap);
 			break;
 		}
 	}
 
+out:
 	srcu_read_unlock(&kvm->srcu, idx);
 
-	return r;
+	return allowed;
 }
 EXPORT_SYMBOL_GPL(kvm_msr_allowed);
 
@@ -2551,6 +2560,8 @@ static void kvm_gen_update_masterclock(struct kvm *kvm)
 	struct kvm_vcpu *vcpu;
 	struct kvm_arch *ka = &kvm->arch;
 
+	kvm_hv_invalidate_tsc_page(kvm);
+
 	spin_lock(&ka->pvclock_gtod_sync_lock);
 	kvm_make_mclock_inprogress_request(kvm);
 	/* no guest entries from this point */
@@ -5352,25 +5363,34 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 	return r;
 }
 
-static void kvm_clear_msr_filter(struct kvm *kvm)
+static struct kvm_x86_msr_filter *kvm_alloc_msr_filter(bool default_allow)
+{
+	struct kvm_x86_msr_filter *msr_filter;
+
+	msr_filter = kzalloc(sizeof(*msr_filter), GFP_KERNEL_ACCOUNT);
+	if (!msr_filter)
+		return NULL;
+
+	msr_filter->default_allow = default_allow;
+	return msr_filter;
+}
+
+static void kvm_free_msr_filter(struct kvm_x86_msr_filter *msr_filter)
 {
 	u32 i;
-	u32 count = kvm->arch.msr_filter.count;
-	struct msr_bitmap_range ranges[16];
 
-	mutex_lock(&kvm->lock);
-	kvm->arch.msr_filter.count = 0;
-	memcpy(ranges, kvm->arch.msr_filter.ranges, count * sizeof(ranges[0]));
-	mutex_unlock(&kvm->lock);
-	synchronize_srcu(&kvm->srcu);
+	if (!msr_filter)
+		return;
 
-	for (i = 0; i < count; i++)
-		kfree(ranges[i].bitmap);
+	for (i = 0; i < msr_filter->count; i++)
+		kfree(msr_filter->ranges[i].bitmap);
+
+	kfree(msr_filter);
 }
 
-static int kvm_add_msr_filter(struct kvm *kvm, struct kvm_msr_filter_range *user_range)
+static int kvm_add_msr_filter(struct kvm_x86_msr_filter *msr_filter,
+			      struct kvm_msr_filter_range *user_range)
 {
-	struct msr_bitmap_range *ranges = kvm->arch.msr_filter.ranges;
 	struct msr_bitmap_range range;
 	unsigned long *bitmap = NULL;
 	size_t bitmap_size;
@@ -5404,11 +5424,9 @@ static int kvm_add_msr_filter(struct kvm *kvm, struct kvm_msr_filter_range *user
 		goto err;
 	}
 
-	/* Everything ok, add this range identifier to our global pool */
-	ranges[kvm->arch.msr_filter.count] = range;
-	/* Make sure we filled the array before we tell anyone to walk it */
-	smp_wmb();
-	kvm->arch.msr_filter.count++;
+	/* Everything ok, add this range identifier. */
+	msr_filter->ranges[msr_filter->count] = range;
+	msr_filter->count++;
 
 	return 0;
 err:
@@ -5419,10 +5437,11 @@ static int kvm_add_msr_filter(struct kvm *kvm, struct kvm_msr_filter_range *user
 static int kvm_vm_ioctl_set_msr_filter(struct kvm *kvm, void __user *argp)
 {
 	struct kvm_msr_filter __user *user_msr_filter = argp;
+	struct kvm_x86_msr_filter *new_filter, *old_filter;
 	struct kvm_msr_filter filter;
 	bool default_allow;
-	int r = 0;
 	bool empty = true;
+	int r = 0;
 	u32 i;
 
 	if (copy_from_user(&filter, user_msr_filter, sizeof(filter)))
@@ -5435,25 +5454,32 @@ static int kvm_vm_ioctl_set_msr_filter(struct kvm *kvm, void __user *argp)
 	if (empty && !default_allow)
 		return -EINVAL;
 
-	kvm_clear_msr_filter(kvm);
-
-	kvm->arch.msr_filter.default_allow = default_allow;
+	new_filter = kvm_alloc_msr_filter(default_allow);
+	if (!new_filter)
+		return -ENOMEM;
 
-	/*
-	 * Protect from concurrent calls to this function that could trigger
-	 * a TOCTOU violation on kvm->arch.msr_filter.count.
-	 */
-	mutex_lock(&kvm->lock);
 	for (i = 0; i < ARRAY_SIZE(filter.ranges); i++) {
-		r = kvm_add_msr_filter(kvm, &filter.ranges[i]);
-		if (r)
-			break;
+		r = kvm_add_msr_filter(new_filter, &filter.ranges[i]);
+		if (r) {
+			kvm_free_msr_filter(new_filter);
+			return r;
+		}
 	}
 
+	mutex_lock(&kvm->lock);
+
+	/* The per-VM filter is protected by kvm->lock... */
+	old_filter = srcu_dereference_check(kvm->arch.msr_filter, &kvm->srcu, 1);
+
+	rcu_assign_pointer(kvm->arch.msr_filter, new_filter);
+	synchronize_srcu(&kvm->srcu);
+
+	kvm_free_msr_filter(old_filter);
+
 	kvm_make_all_cpus_request(kvm, KVM_REQ_MSR_FILTER_CHANGED);
 	mutex_unlock(&kvm->lock);
 
-	return r;
+	return 0;
 }
 
 long kvm_arch_vm_ioctl(struct file *filp,
@@ -6603,7 +6629,7 @@ static int kvm_emulate_wbinvd_noskip(struct kvm_vcpu *vcpu)
 		int cpu = get_cpu();
 
 		cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask);
-		smp_call_function_many(vcpu->arch.wbinvd_dirty_mask,
+		on_each_cpu_mask(vcpu->arch.wbinvd_dirty_mask,
 				wbinvd_ipi, NULL, 1);
 		put_cpu();
 		cpumask_clear(vcpu->arch.wbinvd_dirty_mask);
@@ -10634,8 +10660,6 @@ void kvm_arch_pre_destroy_vm(struct kvm *kvm)
 
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
-	u32 i;
-
 	if (current->mm == kvm->mm) {
 		/*
 		 * Free memory regions allocated on behalf of userspace,
@@ -10651,8 +10675,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 		mutex_unlock(&kvm->slots_lock);
 	}
 	static_call_cond(kvm_x86_vm_destroy)(kvm);
-	for (i = 0; i < kvm->arch.msr_filter.count; i++)
-		kfree(kvm->arch.msr_filter.ranges[i].bitmap);
+	kvm_free_msr_filter(srcu_dereference_check(kvm->arch.msr_filter, &kvm->srcu, 1));
 	kvm_pic_destroy(kvm);
 	kvm_ioapic_destroy(kvm);
 	kvm_free_vcpus(kvm);

tools/testing/selftests/kvm/.gitignore

@@ -8,10 +8,13 @@
 /x86_64/debug_regs
 /x86_64/evmcs_test
 /x86_64/get_cpuid_test
+/x86_64/get_msr_index_features
 /x86_64/kvm_pv_test
+/x86_64/hyperv_clock
 /x86_64/hyperv_cpuid
 /x86_64/mmio_warning_test
 /x86_64/platform_info_test
+/x86_64/set_boot_cpu_id
 /x86_64/set_sregs_test
 /x86_64/smm_test
 /x86_64/state_test

tools/testing/selftests/kvm/Makefile

@@ -39,12 +39,15 @@ LIBKVM_aarch64 = lib/aarch64/processor.c lib/aarch64/ucall.c
 LIBKVM_s390x = lib/s390x/processor.c lib/s390x/ucall.c lib/s390x/diag318_test_handler.c
 
 TEST_GEN_PROGS_x86_64 = x86_64/cr4_cpuid_sync_test
+TEST_GEN_PROGS_x86_64 += x86_64/get_msr_index_features
 TEST_GEN_PROGS_x86_64 += x86_64/evmcs_test
 TEST_GEN_PROGS_x86_64 += x86_64/get_cpuid_test
+TEST_GEN_PROGS_x86_64 += x86_64/hyperv_clock
 TEST_GEN_PROGS_x86_64 += x86_64/hyperv_cpuid
 TEST_GEN_PROGS_x86_64 += x86_64/kvm_pv_test
 TEST_GEN_PROGS_x86_64 += x86_64/mmio_warning_test
 TEST_GEN_PROGS_x86_64 += x86_64/platform_info_test
+TEST_GEN_PROGS_x86_64 += x86_64/set_boot_cpu_id
 TEST_GEN_PROGS_x86_64 += x86_64/set_sregs_test
 TEST_GEN_PROGS_x86_64 += x86_64/smm_test
 TEST_GEN_PROGS_x86_64 += x86_64/state_test

tools/testing/selftests/kvm/include/kvm_util.h

@@ -16,6 +16,7 @@
 
 #include "sparsebit.h"
 
+#define KVM_DEV_PATH "/dev/kvm"
 #define KVM_MAX_VCPUS 512
 
 /*
@@ -133,6 +134,7 @@ void vcpu_ioctl(struct kvm_vm *vm, uint32_t vcpuid, unsigned long ioctl,
 int _vcpu_ioctl(struct kvm_vm *vm, uint32_t vcpuid, unsigned long ioctl,
 		void *arg);
 void vm_ioctl(struct kvm_vm *vm, unsigned long ioctl, void *arg);
+int _vm_ioctl(struct kvm_vm *vm, unsigned long cmd, void *arg);
 void kvm_ioctl(struct kvm_vm *vm, unsigned long ioctl, void *arg);
 int _kvm_ioctl(struct kvm_vm *vm, unsigned long ioctl, void *arg);
 void vm_mem_region_set_flags(struct kvm_vm *vm, uint32_t slot, uint32_t flags);

tools/testing/selftests/kvm/lib/kvm_util.c

@@ -1697,11 +1697,16 @@ void vm_ioctl(struct kvm_vm *vm, unsigned long cmd, void *arg)
 {
 	int ret;
 
-	ret = ioctl(vm->fd, cmd, arg);
+	ret = _vm_ioctl(vm, cmd, arg);
 	TEST_ASSERT(ret == 0, "vm ioctl %lu failed, rc: %i errno: %i (%s)",
 		cmd, ret, errno, strerror(errno));
 }
 
+int _vm_ioctl(struct kvm_vm *vm, unsigned long cmd, void *arg)
+{
+	return ioctl(vm->fd, cmd, arg);
+}
+
 /*
  * KVM system ioctl
  *

tools/testing/selftests/kvm/lib/kvm_util_internal.h

@@ -10,8 +10,6 @@
 
 #include "sparsebit.h"
 
-#define KVM_DEV_PATH		"/dev/kvm"
-
 struct userspace_mem_region {
 	struct kvm_userspace_memory_region region;
 	struct sparsebit *unused_phy_pages;

tools/testing/selftests/kvm/x86_64/get_msr_index_features.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Test that KVM_GET_MSR_INDEX_LIST and
+ * KVM_GET_MSR_FEATURE_INDEX_LIST work as intended
+ *
+ * Copyright (C) 2020, Red Hat, Inc.
+ */
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+static int kvm_num_index_msrs(int kvm_fd, int nmsrs)
+{
+	struct kvm_msr_list *list;
+	int r;
+
+	list = malloc(sizeof(*list) + nmsrs * sizeof(list->indices[0]));
+	list->nmsrs = nmsrs;
+	r = ioctl(kvm_fd, KVM_GET_MSR_INDEX_LIST, list);
+	TEST_ASSERT(r == -1 && errno == E2BIG,
+				"Unexpected result from KVM_GET_MSR_INDEX_LIST probe, r: %i",
+				r);
+
+	r = list->nmsrs;
+	free(list);
+	return r;
+}
+
+static void test_get_msr_index(void)
+{
+	int old_res, res, kvm_fd, r;
+	struct kvm_msr_list *list;
+
+	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
+	if (kvm_fd < 0)
+		exit(KSFT_SKIP);
+
+	old_res = kvm_num_index_msrs(kvm_fd, 0);
+	TEST_ASSERT(old_res != 0, "Expecting nmsrs to be > 0");
+
+	if (old_res != 1) {
+		res = kvm_num_index_msrs(kvm_fd, 1);
+		TEST_ASSERT(res > 1, "Expecting nmsrs to be > 1");
+		TEST_ASSERT(res == old_res, "Expecting nmsrs to be identical");
+	}
+
+	list = malloc(sizeof(*list) + old_res * sizeof(list->indices[0]));
+	list->nmsrs = old_res;
+	r = ioctl(kvm_fd, KVM_GET_MSR_INDEX_LIST, list);
+
+	TEST_ASSERT(r == 0,
+		    "Unexpected result from KVM_GET_MSR_FEATURE_INDEX_LIST, r: %i",
+		    r);
+	TEST_ASSERT(list->nmsrs == old_res, "Expecting nmsrs to be identical");
+	free(list);
+
+	close(kvm_fd);
+}
+
+static int kvm_num_feature_msrs(int kvm_fd, int nmsrs)
+{
+	struct kvm_msr_list *list;
+	int r;
+
+	list = malloc(sizeof(*list) + nmsrs * sizeof(list->indices[0]));
+	list->nmsrs = nmsrs;
+	r = ioctl(kvm_fd, KVM_GET_MSR_FEATURE_INDEX_LIST, list);
+	TEST_ASSERT(r == -1 && errno == E2BIG,
+		"Unexpected result from KVM_GET_MSR_FEATURE_INDEX_LIST probe, r: %i",
+				r);
+
+	r = list->nmsrs;
+	free(list);
+	return r;
+}
+
+struct kvm_msr_list *kvm_get_msr_feature_list(int kvm_fd, int nmsrs)
+{
+	struct kvm_msr_list *list;
+	int r;
+
+	list = malloc(sizeof(*list) + nmsrs * sizeof(list->indices[0]));
+	list->nmsrs = nmsrs;
+	r = ioctl(kvm_fd, KVM_GET_MSR_FEATURE_INDEX_LIST, list);
+
+	TEST_ASSERT(r == 0,
+		"Unexpected result from KVM_GET_MSR_FEATURE_INDEX_LIST, r: %i",
+		r);
+
+	return list;
+}
+
+static void test_get_msr_feature(void)
+{
+	int res, old_res, i, kvm_fd;
+	struct kvm_msr_list *feature_list;
+
+	kvm_fd = open(KVM_DEV_PATH, O_RDONLY);
+	if (kvm_fd < 0)
+		exit(KSFT_SKIP);
+
+	old_res = kvm_num_feature_msrs(kvm_fd, 0);
+	TEST_ASSERT(old_res != 0, "Expecting nmsrs to be > 0");
+
+	if (old_res != 1) {
+		res = kvm_num_feature_msrs(kvm_fd, 1);
+		TEST_ASSERT(res > 1, "Expecting nmsrs to be > 1");
+		TEST_ASSERT(res == old_res, "Expecting nmsrs to be identical");
+	}
+
+	feature_list = kvm_get_msr_feature_list(kvm_fd, old_res);
+	TEST_ASSERT(old_res == feature_list->nmsrs,
+				"Unmatching number of msr indexes");
+
+	for (i = 0; i < feature_list->nmsrs; i++)
+		kvm_get_feature_msr(feature_list->indices[i]);
+
+	free(feature_list);
+	close(kvm_fd);
+}
+
+int main(int argc, char *argv[])
+{
+	if (kvm_check_cap(KVM_CAP_GET_MSR_FEATURES))
+		test_get_msr_feature();
+
+	test_get_msr_index();
+}

tools/testing/selftests/kvm/x86_64/hyperv_clock.c

+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021, Red Hat, Inc.
+ *
+ * Tests for Hyper-V clocksources
+ */
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+struct ms_hyperv_tsc_page {
+	volatile u32 tsc_sequence;
+	u32 reserved1;
+	volatile u64 tsc_scale;
+	volatile s64 tsc_offset;
+} __packed;
+
+#define HV_X64_MSR_GUEST_OS_ID			0x40000000
+#define HV_X64_MSR_TIME_REF_COUNT		0x40000020
+#define HV_X64_MSR_REFERENCE_TSC		0x40000021
+#define HV_X64_MSR_TSC_FREQUENCY		0x40000022
+#define HV_X64_MSR_REENLIGHTENMENT_CONTROL	0x40000106
+#define HV_X64_MSR_TSC_EMULATION_CONTROL	0x40000107
+
+/* Simplified mul_u64_u64_shr() */
+static inline u64 mul_u64_u64_shr64(u64 a, u64 b)
+{
+	union {
+		u64 ll;
+		struct {
+			u32 low, high;
+		} l;
+	} rm, rn, rh, a0, b0;
+	u64 c;
+
+	a0.ll = a;
+	b0.ll = b;
+
+	rm.ll = (u64)a0.l.low * b0.l.high;
+	rn.ll = (u64)a0.l.high * b0.l.low;
+	rh.ll = (u64)a0.l.high * b0.l.high;
+
+	rh.l.low = c = rm.l.high + rn.l.high + rh.l.low;
+	rh.l.high = (c >> 32) + rh.l.high;
+
+	return rh.ll;
+}
+
+static inline void nop_loop(void)
+{
+	int i;
+
+	for (i = 0; i < 1000000; i++)
+		asm volatile("nop");
+}
+
+static inline void check_tsc_msr_rdtsc(void)
+{
+	u64 tsc_freq, r1, r2, t1, t2;
+	s64 delta_ns;
+
+	tsc_freq = rdmsr(HV_X64_MSR_TSC_FREQUENCY);
+	GUEST_ASSERT(tsc_freq > 0);
+
+	/* First, check MSR-based clocksource */
+	r1 = rdtsc();
+	t1 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
+	nop_loop();
+	r2 = rdtsc();
+	t2 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
+
+	GUEST_ASSERT(r2 > r1 && t2 > t1);
+
+	/* HV_X64_MSR_TIME_REF_COUNT is in 100ns */
+	delta_ns = ((t2 - t1) * 100) - ((r2 - r1) * 1000000000 / tsc_freq);
+	if (delta_ns < 0)
+		delta_ns = -delta_ns;
+
+	/* 1% tolerance */
+	GUEST_ASSERT(delta_ns * 100 < (t2 - t1) * 100);
+}
+
+static inline void check_tsc_msr_tsc_page(struct ms_hyperv_tsc_page *tsc_page)
+{
+	u64 r1, r2, t1, t2;
+
+	/* Compare TSC page clocksource with HV_X64_MSR_TIME_REF_COUNT */
+	t1 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
+	r1 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
+
+	/* 10 ms tolerance */
+	GUEST_ASSERT(r1 >= t1 && r1 - t1 < 100000);
+	nop_loop();
+
+	t2 = mul_u64_u64_shr64(rdtsc(), tsc_page->tsc_scale) + tsc_page->tsc_offset;
+	r2 = rdmsr(HV_X64_MSR_TIME_REF_COUNT);
+	GUEST_ASSERT(r2 >= t1 && r2 - t2 < 100000);
+}
+
+static void guest_main(struct ms_hyperv_tsc_page *tsc_page, vm_paddr_t tsc_page_gpa)
+{
+	u64 tsc_scale, tsc_offset;
+
+	/* Set Guest OS id to enable Hyper-V emulation */
+	GUEST_SYNC(1);
+	wrmsr(HV_X64_MSR_GUEST_OS_ID, (u64)0x8100 << 48);
+	GUEST_SYNC(2);
+
+	check_tsc_msr_rdtsc();
+
+	GUEST_SYNC(3);
+
+	/* Set up TSC page is disabled state, check that it's clean */
+	wrmsr(HV_X64_MSR_REFERENCE_TSC, tsc_page_gpa);
+	GUEST_ASSERT(tsc_page->tsc_sequence == 0);
+	GUEST_ASSERT(tsc_page->tsc_scale == 0);
+	GUEST_ASSERT(tsc_page->tsc_offset == 0);
+
+	GUEST_SYNC(4);
+
+	/* Set up TSC page is enabled state */
+	wrmsr(HV_X64_MSR_REFERENCE_TSC, tsc_page_gpa | 0x1);
+	GUEST_ASSERT(tsc_page->tsc_sequence != 0);
+
+	GUEST_SYNC(5);
+
+	check_tsc_msr_tsc_page(tsc_page);
+
+	GUEST_SYNC(6);
+
+	tsc_offset = tsc_page->tsc_offset;
+	/* Call KVM_SET_CLOCK from userspace, check that TSC page was updated */
+	GUEST_SYNC(7);
+	GUEST_ASSERT(tsc_page->tsc_offset != tsc_offset);
+
+	nop_loop();
+
+	/*
+	 * Enable Re-enlightenment and check that TSC page stays constant across
+	 * KVM_SET_CLOCK.
+	 */
+	wrmsr(HV_X64_MSR_REENLIGHTENMENT_CONTROL, 0x1 << 16 | 0xff);
+	wrmsr(HV_X64_MSR_TSC_EMULATION_CONTROL, 0x1);
+	tsc_offset = tsc_page->tsc_offset;
+	tsc_scale = tsc_page->tsc_scale;
+	GUEST_SYNC(8);
+	GUEST_ASSERT(tsc_page->tsc_offset == tsc_offset);
+	GUEST_ASSERT(tsc_page->tsc_scale == tsc_scale);
+
+	GUEST_SYNC(9);
+
+	check_tsc_msr_tsc_page(tsc_page);
+
+	/*
+	 * Disable re-enlightenment and TSC page, check that KVM doesn't update
+	 * it anymore.
+	 */
+	wrmsr(HV_X64_MSR_REENLIGHTENMENT_CONTROL, 0);
+	wrmsr(HV_X64_MSR_TSC_EMULATION_CONTROL, 0);
+	wrmsr(HV_X64_MSR_REFERENCE_TSC, 0);
+	memset(tsc_page, 0, sizeof(*tsc_page));
+
+	GUEST_SYNC(10);
+	GUEST_ASSERT(tsc_page->tsc_sequence == 0);
+	GUEST_ASSERT(tsc_page->tsc_offset == 0);
+	GUEST_ASSERT(tsc_page->tsc_scale == 0);
+
+	GUEST_DONE();
+}
+
+#define VCPU_ID 0
+
+static void host_check_tsc_msr_rdtsc(struct kvm_vm *vm)
+{
+	u64 tsc_freq, r1, r2, t1, t2;
+	s64 delta_ns;
+
+	tsc_freq = vcpu_get_msr(vm, VCPU_ID, HV_X64_MSR_TSC_FREQUENCY);
+	TEST_ASSERT(tsc_freq > 0, "TSC frequency must be nonzero");
+
+	/* First, check MSR-based clocksource */
+	r1 = rdtsc();
+	t1 = vcpu_get_msr(vm, VCPU_ID, HV_X64_MSR_TIME_REF_COUNT);
+	nop_loop();
+	r2 = rdtsc();
+	t2 = vcpu_get_msr(vm, VCPU_ID, HV_X64_MSR_TIME_REF_COUNT);
+
+	TEST_ASSERT(t2 > t1, "Time reference MSR is not monotonic (%ld <= %ld)", t1, t2);
+
+	/* HV_X64_MSR_TIME_REF_COUNT is in 100ns */
+	delta_ns = ((t2 - t1) * 100) - ((r2 - r1) * 1000000000 / tsc_freq);
+	if (delta_ns < 0)
+		delta_ns = -delta_ns;
+
+	/* 1% tolerance */
+	TEST_ASSERT(delta_ns * 100 < (t2 - t1) * 100,
+		    "Elapsed time does not match (MSR=%ld, TSC=%ld)",
+		    (t2 - t1) * 100, (r2 - r1) * 1000000000 / tsc_freq);
+}
+
+int main(void)
+{
+	struct kvm_vm *vm;
+	struct kvm_run *run;
+	struct ucall uc;
+	vm_vaddr_t tsc_page_gva;
+	int stage;
+
+	vm = vm_create_default(VCPU_ID, 0, guest_main);
+	run = vcpu_state(vm, VCPU_ID);
+
+	vcpu_set_hv_cpuid(vm, VCPU_ID);
+
+	tsc_page_gva = vm_vaddr_alloc(vm, getpagesize(), 0x10000, 0, 0);
+	memset(addr_gpa2hva(vm, tsc_page_gva), 0x0, getpagesize());
+	TEST_ASSERT((addr_gva2gpa(vm, tsc_page_gva) & (getpagesize() - 1)) == 0,
+		"TSC page has to be page aligned\n");
+	vcpu_args_set(vm, VCPU_ID, 2, tsc_page_gva, addr_gva2gpa(vm, tsc_page_gva));
+
+	host_check_tsc_msr_rdtsc(vm);
+
+	for (stage = 1;; stage++) {
+		_vcpu_run(vm, VCPU_ID);
+		TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
+			    "Stage %d: unexpected exit reason: %u (%s),\n",
+			    stage, run->exit_reason,
+			    exit_reason_str(run->exit_reason));
+
+		switch (get_ucall(vm, VCPU_ID, &uc)) {
+		case UCALL_ABORT:
+			TEST_FAIL("%s at %s:%ld", (const char *)uc.args[0],
+				  __FILE__, uc.args[1]);
+			/* NOT REACHED */
+		case UCALL_SYNC:
+			break;
+		case UCALL_DONE:
+			/* Keep in sync with guest_main() */
+			TEST_ASSERT(stage == 11, "Testing ended prematurely, stage %d\n",
+				    stage);
+			goto out;
+		default:
+			TEST_FAIL("Unknown ucall %lu", uc.cmd);
+		}
+
+		TEST_ASSERT(!strcmp((const char *)uc.args[0], "hello") &&
+			    uc.args[1] == stage,
+			    "Stage %d: Unexpected register values vmexit, got %lx",
+			    stage, (ulong)uc.args[1]);
+
+		/* Reset kvmclock triggering TSC page update */
+		if (stage == 7 || stage == 8 || stage == 10) {
+			struct kvm_clock_data clock = {0};
+
+			vm_ioctl(vm, KVM_SET_CLOCK, &clock);
+		}
+	}
+
+out:
+	kvm_vm_free(vm);
+}

tools/testing/selftests/kvm/x86_64/set_boot_cpu_id.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Test that KVM_SET_BOOT_CPU_ID works as intended
+ *
+ * Copyright (C) 2020, Red Hat, Inc.
+ */
+#define _GNU_SOURCE /* for program_invocation_name */
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+
+#define N_VCPU 2
+#define VCPU_ID0 0
+#define VCPU_ID1 1
+
+static uint32_t get_bsp_flag(void)
+{
+	return rdmsr(MSR_IA32_APICBASE) & MSR_IA32_APICBASE_BSP;
+}
+
+static void guest_bsp_vcpu(void *arg)
+{
+	GUEST_SYNC(1);
+
+	GUEST_ASSERT(get_bsp_flag() != 0);
+
+	GUEST_DONE();
+}
+
+static void guest_not_bsp_vcpu(void *arg)
+{
+	GUEST_SYNC(1);
+
+	GUEST_ASSERT(get_bsp_flag() == 0);
+
+	GUEST_DONE();
+}
+
+static void test_set_boot_busy(struct kvm_vm *vm)
+{
+	int res;
+
+	res = _vm_ioctl(vm, KVM_SET_BOOT_CPU_ID, (void *) VCPU_ID0);
+	TEST_ASSERT(res == -1 && errno == EBUSY,
+			"KVM_SET_BOOT_CPU_ID set while running vm");
+}
+
+static void run_vcpu(struct kvm_vm *vm, uint32_t vcpuid)
+{
+	struct ucall uc;
+	int stage;
+
+	for (stage = 0; stage < 2; stage++) {
+
+		vcpu_run(vm, vcpuid);
+
+		switch (get_ucall(vm, vcpuid, &uc)) {
+		case UCALL_SYNC:
+			TEST_ASSERT(!strcmp((const char *)uc.args[0], "hello") &&
+					uc.args[1] == stage + 1,
+					"Stage %d: Unexpected register values vmexit, got %lx",
+					stage + 1, (ulong)uc.args[1]);
+			test_set_boot_busy(vm);
+			break;
+		case UCALL_DONE:
+			TEST_ASSERT(stage == 1,
+					"Expected GUEST_DONE in stage 2, got stage %d",
+					stage);
+			break;
+		case UCALL_ABORT:
+			TEST_ASSERT(false, "%s at %s:%ld\n\tvalues: %#lx, %#lx",
+						(const char *)uc.args[0], __FILE__,
+						uc.args[1], uc.args[2], uc.args[3]);
+		default:
+			TEST_ASSERT(false, "Unexpected exit: %s",
+					exit_reason_str(vcpu_state(vm, vcpuid)->exit_reason));
+		}
+	}
+}
+
+static struct kvm_vm *create_vm(void)
+{
+	struct kvm_vm *vm;
+	uint64_t vcpu_pages = (DEFAULT_STACK_PGS) * 2;
+	uint64_t extra_pg_pages = vcpu_pages / PTES_PER_MIN_PAGE * N_VCPU;
+	uint64_t pages = DEFAULT_GUEST_PHY_PAGES + vcpu_pages + extra_pg_pages;
+
+	pages = vm_adjust_num_guest_pages(VM_MODE_DEFAULT, pages);
+	vm = vm_create(VM_MODE_DEFAULT, pages, O_RDWR);
+
+	kvm_vm_elf_load(vm, program_invocation_name, 0, 0);
+	vm_create_irqchip(vm);
+
+	return vm;
+}
+
+static void add_x86_vcpu(struct kvm_vm *vm, uint32_t vcpuid, bool bsp_code)
+{
+	if (bsp_code)
+		vm_vcpu_add_default(vm, vcpuid, guest_bsp_vcpu);
+	else
+		vm_vcpu_add_default(vm, vcpuid, guest_not_bsp_vcpu);
+
+	vcpu_set_cpuid(vm, vcpuid, kvm_get_supported_cpuid());
+}
+
+static void run_vm_bsp(uint32_t bsp_vcpu)
+{
+	struct kvm_vm *vm;
+	bool is_bsp_vcpu1 = bsp_vcpu == VCPU_ID1;
+
+	vm = create_vm();
+
+	if (is_bsp_vcpu1)
+		vm_ioctl(vm, KVM_SET_BOOT_CPU_ID, (void *) VCPU_ID1);
+
+	add_x86_vcpu(vm, VCPU_ID0, !is_bsp_vcpu1);
+	add_x86_vcpu(vm, VCPU_ID1, is_bsp_vcpu1);
+
+	run_vcpu(vm, VCPU_ID0);
+	run_vcpu(vm, VCPU_ID1);
+
+	kvm_vm_free(vm);
+}
+
+static void check_set_bsp_busy(void)
+{
+	struct kvm_vm *vm;
+	int res;
+
+	vm = create_vm();
+
+	add_x86_vcpu(vm, VCPU_ID0, true);
+	add_x86_vcpu(vm, VCPU_ID1, false);
+
+	res = _vm_ioctl(vm, KVM_SET_BOOT_CPU_ID, (void *) VCPU_ID1);
+	TEST_ASSERT(res == -1 && errno == EBUSY, "KVM_SET_BOOT_CPU_ID set after adding vcpu");
+
+	run_vcpu(vm, VCPU_ID0);
+	run_vcpu(vm, VCPU_ID1);
+
+	res = _vm_ioctl(vm, KVM_SET_BOOT_CPU_ID, (void *) VCPU_ID1);
+	TEST_ASSERT(res == -1 && errno == EBUSY, "KVM_SET_BOOT_CPU_ID set to a terminated vcpu");
+
+	kvm_vm_free(vm);
+}
+
+int main(int argc, char *argv[])
+{
+	if (!kvm_check_cap(KVM_CAP_SET_BOOT_CPU_ID)) {
+		print_skip("set_boot_cpu_id not available");
+		return 0;
+	}
+
+	run_vm_bsp(VCPU_ID0);
+	run_vm_bsp(VCPU_ID1);
+	run_vm_bsp(VCPU_ID0);
+
+	check_set_bsp_busy();
+}